In order to understand the latency of human immunodeficiency virus (HIV-1) infection and the effect of various cofactors, several experimental models have been developed. Although some non-lymphoid cells are not usually considered as the main target tissue for HIV-1 infection, they can serve as important models. Together with other HIV-1 permissive cellular models we have also used human hepatoblastoma HepG2 cells, since hepatic abnormalities and a higher incidence of hepatitis B virus infection are found with AIDS. In contrast to the stimulatory effect of tumor necrosis factor (TNF-alpha) or Phorbol-12-myristate-13-acetate (PMA) in HIV-1 replication in T cells, these agents inhibited HIV-1 replication in liver cells. The long-term objective and specific aims of this proposal are the following: 1) Determine the mechanism(s) involved in inhibition of HIV-1 infection by TNF-alpha in HepG2 cells. Compare the effect of TNF-alpha with that of PMA on HIV-1 infection in this system, which may indicate a novel pathway for HIV-1 infection. 2) Compare these data with other cell lines including the HepG2 clone which is CD4 negative. 3) Analyze the state of the HIV-1 DNA and RNA in these infected cells. 4) Isolate and characterize TNF-alpha and PMA-induced gene sequences by subtraction hybridization of HepG2 derived cDNA libraries. These cDNAs will be cloned in a eukaryotic expression vector so that they can be transfected into various cell lines including HepG2 to obtain stable cell lines for analysis of their resistance to HIV-1 infection. 5) We will use various HIV-1 proviral clones for infection, and stable cell lines containing the proviral genome will be isolated. The effect of TNF-alpha or PMA will be evaluated in these infectious clones derived from various hepatoma cell lines. 6) The role of protein kinase C and cAMP in relation to TNF-alpha and PMA effect will be studied. 7) Gel retardation assays and extensive DNA footprinting will be used to identify the proteins in TNF-alpha or PMA-treated cells which bind to the regulatory regions of HIV-1 and TAR RNA sequences. 8) By mutating various regions of this proviral clone, we intend to localize the region(s) in HIV-1 genome responsible for the TNF- alpha or PMA mediated inhibition. 9) We will identify, characterize, purify, and, if required, isolate the cDNAs encoding the trans-acting proteins from various hepatoma and other cell lines that bind to HIV-1 LTR by screening lambdagt11 library and compare with that of treated cells.